• The Presence of HLA-B*5701, -DRB1*0701, and -DQ3 Is Highly Predictive of Hypersensitivity to the HIV Reverse Transcriptase Inhibitor Abacavir
  Abstract 91
  Authored by S. Mallal¹, D. Nolan¹, C. Witt², G. Masel^1, 2, A. Martin¹, C. Moore¹, D. Sayer^1, 2, A. Castley², C. Mamotte^1, 2, D. Maxwell¹, I. James¹, and F. Christiansen² (¹Ctr. for Clin. Immunology and Biomed. Statistics, Royal Perth Hosp., Australia and ²Univ. of Western Australia, Murdoch, Perth, Australia)
  View the original abstract
• HLA-B57 and TNF-Alpha Variants Associated With Hypersensitivity Reactions to Abacavir Among HIV-1-Positive Subjects
  Abstract 92
  Authored by S. Hetherington, A. Hughes, M. Mosteller, D. Shortino, K. Baker, E. Lai, M. Stocum, and A. Roses (GlaxoSmithKline, Res. Triangle Park, NC)
  View the original abstract

The Presence of HLA-B*5701, -DRB1*0701, and -DQ3 Is Highly Predictive of Hypersensitivity to the HIV Reverse Transcriptase Inhibitor Abacavir (Abstract 91)

S. Mallal and colleagues from the Royal Perth Hospital in Perth, Australia evaluated the association of certain ancestral haplotypes with the abacavir hypersensitivity reaction (AHR). Abacavir hypersensitivity reaction occurs in approximately five percent of persons started on abacavir and can be life threatening. Abacavir hypersensitivity reaction is usually seen within six weeks of beginning abacavir and is characterized by two of the following: fever, rash, GI symptoms, lethargy, malaise, myalgias and respiratory complaints. The abacavir hypersensitivity reaction usually resolves within 72 hours of stopping abacavir. If abacavir hypersensitivity reaction is suspected, a patient should never be restarted on abacavir.

A genetic link with abacavir hypersensitivity reaction was pursued because of observations that abacavir hypersensitivity reaction has occurred in family members (familial predisposition), the rate of abacavir hypersensitivity reaction appears to be lower in African Americans, and other multi-system hypersensitivity drug reactions have involved genetic associations.

The first 200 patients taking abacavir in the Western Australian HIV Cohort Study underwent major histocompatability testing. Using strict clinical criteria, abacavir hypersensitivity reaction was diagnosed in 18 of the 200. HLA-B57 was present in 78 percent of patients who had an abacavir hypersensitivity reaction and 2.3 percent of abacavir tolerant subjects (odds ratio = 117, p<0.0001). The HLA-DRB1 and -DQ3 combination was found in 72 percent of those who had an abacavir hypersensitivity reaction and in none of the abacavir tolerant subjects. If prospective haplotyping had been performed and abacavir withheld from persons positive for HLA-B57, -DRB1, and -DQ3, then the rate of hypersensitivity would have dropped from 9 percent to 2.5 percent.

Importantly, testing for the haplotype is not 100 percent sensitive and cannot be considered a screening test. Management of abacavir hypersensitivity reaction still requires clinical awareness. This type of testing also could not be used to decide whether to re-challenge with abacavir. The authors stressed that this linkage was noted only for a mostly Caucasian population in Western Australia. This data was recently published in The Lancet (2002; 359: 727-732). This study is an interesting approach to the evolving field of pharmacogenomics but can't be applied to clinical practice at the present time.

HLA-B57 and TNF-Alpha Variants Associated With Hypersensitivity Reactions to Abacavir Among HIV-1-Positive Subjects (Abstract 92)

S. Hetherington and colleagues from GlaxoSmithKline in Research Park, NC also were interested in the genetics of abacavir hypersensitivity reaction (AHR). They are approaching their work with pharmacogenetics in two phases:

• Phase I -- evaluate among 114 gene markers and HLA A, B or DR.

• Phase II -- study single nucleotide polymorphisms (SNPs) to refine the associations.

This study describes their work with Phase I. Eighty-five abacavir hypersensitivity reaction cases were identified along with 115 matched controls who were abacavir tolerant for at least six weeks. Overall, 92 percent of subjects were male and 74 percent Caucasian. Analysis identified a known polymorphism of TNF-alpha (-238A) among 25 of 58 (43 percent) of cases compared to 7 percent of controls (p<0.001). HLA-B57 was present in 39 of 84 (46 percent) of abacavir hypersensitivity reaction subjects versus 4 of 113 controls (3.5 percent; p<0.001). The authors emphasized that their study involved mostly Caucasians and that their modeling of the genetic linkages is incomplete. In the end, the following implications of their data were emphasized:

1. no change in the current clinical approach to abacavir hypersensitivity reaction should be made;

2. there is a high false negative rate so that the genetic tests cannot be used to predict who will develop abacavir hypersensitivity reaction; and

3. the HLA type cannot be used to verify a clinical diagnosis.

Like the paper by Mallal et al., this study is a voyage into the field of pharmacogenomics. For the present, these papers are a proof-of-concept endeavor that cannot be translated into clinical practice but perhaps gives us a glimpse into the future of pharmacogenetics.